Imagebased Tree Branch Recovery Yilin Wang Rulebased using
Image-based Tree Branch Recovery Yilin Wang
Rule-based ◦ using a small set of generative rules or grammar to create branches and leaves Image-based ◦ reconstructing the tree directly from image samples Tree modeling
A L-system consists of an axiom and productions Example L-system
P. Prusinkiewicz, M. James, and R. Mech. Synthetic topiary. In SIGGRAPH ’ 94: Proceedings of the 21 st annual conference on Computer graphics and interactive techniques, pages 351– 358, New York, NY, USA, 1994. ACM. Sample tree structures generated using L-system
P. Tan, G. Zeng, J. Wang, S. B. Kang, and L. Quan. Image-based tree modeling. In SIGGRAPH ’ 07: ACM SIGGRAPH 2007 papers, page 87, New York, NY, USA, 2007. ACM. Image-based tree modeling
P. Tan, G. Zeng, J. Wang, S. B. Kang, and L. Quan. Image-based tree modeling. In SIGGRAPH ’ 07: ACM SIGGRAPH 2007 papers, page 87, New York, NY, USA, 2007. ACM. Bare tree example
Problem ◦ “it is not easy to generate complete tree models from just 3 D points because of the difficulties in determining what is missing and in filling the missing information” Can we extract more useful information from images to refine the branch reconstruction? Motivation
Final goal ◦ To recover the tree branches from images, and to make the procedure as automatic as possible Subtasks ◦ 3 D structure generation from source images ◦ Reconstruction of trunk and visible branches ◦ Recovery of occlude branches Potential improvements ◦ Using segmentations and gradient maps to refine branches, e. g. estimate length, density, and direction of the branch My project
Thank you!
- Slides: 9